Thermal Conductivity of Biological Fibers: Hair, Silk, Cellulose and the Moisture Problem

Biological fibers are among the most-touched materials on Earth and among the least-honestly measured. Hair, wool, silk and cellulose all conduct heat through hierarchical protein or polysaccharide structures whose axial order resembles a drawn polymer — and all of them are hygroscopic, meaning the specimen on the stage exchanges water with the room while you measure it¹². Sorbed water conducts better than air and fills voids that were carrying almost nothing, while also raising heat capacity and swelling the fiber — so a bio-fiber’s measured properties can move substantially between a dry-box morning and a humid afternoon, on the same specimen, with no instrument fault at all. Which way the diffusivity moves is not universal: it depends on how fast conductivity, volumetric heat capacity and geometry each respond in that species. This article treats bio-fibers as what they are — anisotropic, hierarchical, moisture-coupled composites — and covers the structure that carries their heat, the humidity discipline that makes numbers comparable, and the single-fiber methods that can measure a specimen too thin to see.

Structure: why bio-fibers conduct along themselves

Nature builds fibers the way a drawing machine does — by aligning long molecules along an axis — and the thermal consequence is the same. In hair and wool, keratin intermediate filaments run axially through a cortical matrix, sheathed by an overlapping cuticle. In silk, β-sheet nanocrystallites are embedded in a semi-amorphous matrix with pronounced axial texture. In plant fibers, crystalline cellulose microfibrils spiral along the cell wall at characteristic angles. Each architecture gives phonon-like vibrational energy a preferred direction: longer covalent runs along the fiber, weaker inter-chain hops across it — the same competition that governs synthetic polymer fibers, with anisotropy inherited from biology instead of a draw ratio¹². Absolute values do not collapse to one insulating band. Nonporous cellulose fiber and nanofiber networks measure near 0.54–0.57 W m⁻¹K⁻¹, and oriented nanocellulose papers report in-plane values of roughly 0.6–2.5 against through-plane 0.3–0.5 — anisotropy from fiber alignment rather than from chemistry³⁴. At the other extreme, single spider silk has been reported with exceptionally high axial conductivity that rises further under stretching — a direct demonstration that biological polymers can leave the insulating band entirely when their molecular order is high enough⁵. The transferable lesson is that structure, not the label “biological,” sets the number.

The moisture problem

Now add water. Keratin, silk fibroin and cellulose are all hygroscopic: at ordinary room humidity they hold several percent of their mass as sorbed water, rising steeply toward saturation². That water changes the measurement three ways at once. It adds a conducting phase — liquid water at ≈0.6 W m⁻¹K⁻¹ replacing air-filled voids and inter-fibrillar gaps that were the structure’s worst bottlenecks. It raises volumetric heat capacity, since water’s specific heat exceeds that of most organic solids — which matters doubly for transient methods, whose native product α = k/ρc_p depends on the ratio of two quantities that are both rising. Whether α ends up higher or lower is therefore species- and structure-dependent rather than a universal rule, and a diffusivity-based measurement and a conductivity-based measurement can report opposite humidity trends on the same material⁶. And it changes the structure itself, plasticizing the matrix and swelling the fiber’s cross-section — so the geometry entered into the model drifts along with the properties. None of this is exotic; all of it is invisible if the report omits humidity, which is why cross-laboratory comparisons of bio-fiber conductivity so often disagree for entirely non-instrumental reasons⁵.

Interactive: how humidity moves the number

The simulator makes the coupling visible — and deliberately refuses to prejudge its sign. Conductivity climbs as sorbed water fills the bottlenecks; volumetric heat capacity climbs too; and the diffusivity that transient instruments actually measure is their ratio. The second slider sets how strongly capacity responds relative to conductivity in your material, and the readout reports which way α is moving at that setting. Both regimes are physically reachable, which is precisely why a bio-fiber number without a stated humidity is unusable.

The curves are schematic mixture-rule shapes for teaching, not sorption isotherms for any species: real fibers have material-specific uptake curves, hysteresis between wetting and drying, and swelling that shifts the cross-section entering the model. The transferable lesson is the coupling and the reporting habit it demands — not a universal direction for α.

Measuring a single bio-fiber

A hair is tens of micrometers across and electrically insulating — two facts that eliminate most of the instrument catalog. The workable route is the suspended electrothermal family: mount the fiber between electrodes, and because it cannot carry current itself, apply a thin conductive coating that acts as heater and thermometer while a differential protocol subtracts its contribution — a composite treatment rather than a simple subtraction, since the coating adds both conductance and heat capacity⁷⁸. Two additional line items are bio-specific. Vacuum versus humidity is a genuine conflict: suspended thermal measurements use vacuum to suppress residual-gas heat transfer below the selected measurement resolution, but pumping dries the specimen and changes the property you came to measure — so protocols either condition and seal, measure the dry state deliberately, or characterize the drying transient itself and report accordingly⁹¹⁰. Geometry is uncertain: bio-fibers are elliptical, tapered and rough; cross-sectional area — entering the result linearly — deserves microscopy rather than a nominal diameter¹¹. Optothermal and microdevice routes provide useful cross-checks on the smallest specimens; agreement across families with largely independent systematics is strong corroboration rather than proof, since shared assumptions (geometry, emissivity, contact) can survive a method change¹²¹³.

Reporting bio-fiber data that others can reproduce

Five fields make a bio-fiber measurement reproducible: species and provenance (a hair is not a hair — treatment, age and grooming all matter), conditioning state (relative humidity and equilibration time, or explicitly dried and how), direction (axial versus radial, never bare), geometry method (how cross-section was determined), and native quantity plus method (k or α, and which technique produced it)⁶¹¹. Missing any of them, a bio-fiber number is a snapshot of an unspecified afternoon. Carrying all five, it becomes a data point others can build on — the standard our datasheet guide asks of every thermal report and that the thermal testing team applies to biological specimens as a matter of course.

Frequently asked questions

Do all biological fibers conduct about the same?

No — and assuming so is the most common error. Cellulose networks measure near 0.5 W m⁻¹K⁻¹ with modest orientation-driven anisotropy³⁴, while single spider silk has been reported far above that band and to change under stretching⁵. Species, hierarchy and moisture state each move the answer, so quoted values need all three attached.

Does a wet fiber conduct better?

Conductivity generally rises with sorbed water, because water replaces air in the voids. Diffusivity may rise or fall, since volumetric heat capacity rises as well and α is their ratio — the direction depends on which responds faster in that material. Report the conditioning and both quantities rather than assuming a trend⁶.

Why does vacuum complicate biological measurements?

Because the vacuum that removes parasitic gas conduction also dries the specimen. Protocols must choose — and state — whether they report the conditioned or the dried state⁹¹⁰.

Can natural fibers be engineered for higher conductivity?

Increasing axial order is the same lever that works on synthetics, and nanocellulose papers show the orientation dependence directly⁴. Silk demonstrates how far the effect can reach in a biological polymer, including a reported response to stretching⁵. Chemistry, crystallinity and defect population set the ceiling alongside orientation.

Which method should I use for a natural-fiber study?

Coated suspended electrothermal transients for single fibers, with humidity conditioning documented; bulk methods for fabrics and mats, where porosity and air become part of the measured system. The selector formalizes the branch.

Keep Exploring the ACS Thermal Metrology Knowledge Hub

This article is one chapter of the ACS thermal metrology knowledge hub. To keep going:

• Thermal conductivity & diffusivity testing: the pillar guide — methods, samples and a buyer’s framework in one place.
• Polymer fiber thermal conductivity — how drawing turns insulators into conductors.
• The transient electro-thermal technique: a complete guide — the suspended-sample family at the heart of this hub.
• Testing non-conductive samples: the metal-coating protocol — making insulators measurable, then subtracting the film.
• Why thermal measurements need vacuum — convection, radiation and the parasitics they hide.
• Reading thermal datasheets — units, conversions and uncertainty without the traps.
• ACS thermal testing services — the team that turns these distinctions into data on your specimen.

This article discusses the thermal conductivity of biological fibers for educational purposes. Quoted magnitudes are values reported in the cited literature for specific materials and depend strongly on species, treatment, geometry determination and moisture state; the moisture simulator uses schematic mixture-rule shapes rather than species-specific sorption isotherms, and deliberately does not assert a universal direction for the diffusivity trend. For conditioned, direction-resolved measurements on your specimens, contact our thermal testing team.